ARTICLE IN PRESS

0360-5442/$ - se

doi:10.1016/j.en

�CorrespondE-mail addr

Energy 32 (2007) 1448–1454

www.elsevier.com/locate/energy

Technology and petroleum exhaustion: Evidence fromtwo mega-oilfields

John Gowdy�, Roxana Julia

Department of Economics, Rensselaer Polytechnic Institute, 110 8th St., Troy, NY 12180,USA

Received 21 November 2005

Abstract

In this paper, we use results from the Hotelling model of non-renewable resources to examine the mainstream view among economists

that improvements in recovery technology can offset declines in petroleum reserves. We present empirical evidence from two well-

documented mega-oilfields: the Forties in the North Sea and the Yates in West Texas. Patterns of depletion in these two fields suggest

that technology temporarily increases the rates of production at the expense of more pronounced rates of depletion in later years—in line

with Hotelling’s predictions. Insofar as our results are generalizable, they call into question the view of most economists that technology

can mitigate absolute resource scarcity. This raises concerns about the capacity of current mega-fields to meet future oil demand.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Non-renewable resource scarcity; Technological change; Oil depletion

1. The economic view of resource scarcity

A recurring theme in resource economics textbooks isthat absolute resource scarcity is not a major concern foreconomic policy. The standard view is that the pricesystem, by encouraging substitution, exploration, andtechnological advances, in effect creates more resourcesas prices increase. In this view, scarcity is properly seen asan economic not a physical concept [1]. The view expressedin a recent text by Hanley et al. [2] is typical:

As a resource gets scarcer, its price will, other thingsbeing equal, tend to rise. This will reduce consumption(by substitution, for example), and increase production-yWhat is more, as prices rise producers will beencouraged to engage in more exploration, which willincrease the resource base if finds are made.

On the face of it, this argument is quite reasonable.However, it is one thing to accept the general statementthat as a resource becomes scarce its price will rise andsuppliers can profitably search for more of it and invest in

e front matter r 2006 Elsevier Ltd. All rights reserved.

ergy.2006.10.019

ing author. Tel.: +1518 276 8094; fax: +1 518 276 2235.

ess: gowdyj@rpi.edu (J. Gowdy).

technologies that boost production, but quite another toargue that the increasing scarcity of an essential resourcelike petroleum can be easily accommodated. The laterposition has been the basic message of resource economistsfor several decades now. The standard economic position isthat scarcity is a relative, not absolute, concept and thatthere is nothing unique about any particular productiveinput, including petroleum.Recent theoretical developments and empirical evidence

suggests that economists may have been too quick todismiss absolute scarcity. First of all it has become clearthat the Walrasian general equilibrium model, uponwhich standard resource economics is based, dependson some strong assumptions that do not accuratelydescribe real-world consumer or producer behavior [3–6].Theoretical economic models that address the issue ofresource scarcity assume rational and fully informedagents, efficient allocation through time, and that presentand future resource stocks are known. Secondly, theuse of prices as indicators of absolute scarcity has beenshown to be seriously misleading. Much of the empiricalliterature using historical prices to show that resourcesare not becoming scarce does not adhere to the samepremises as the theoretical literature [7,8]. Thirdly, the

ARTICLE IN PRESS

Time to Exhaustion

Price

Original Price Path

C

C'

A Fall

In Cost

TT'

Effect of an increase in

technology resulting in

a fall in extraction costs

P

P'

Fig. 1. Hotelling on technological change that lowers cost. Adapted from

Pearce and Turner [28].

J. Gowdy, R. Julia / Energy 32 (2007) 1448–1454 1449

Ceteris paribus assumption never holds so that wehave to worry about a number complicating factorsalways present when dealing with non-renewableresources. These include the Jevons effect that efficiencyleads to lower prices and more consumption [9],the presence of historical lock-in of built infr-astructure, and the political power of vested interestsdependent upon continued exploitation of a particularresource [10], just to name a few. Finally, economists tendto see technology as a free good that depends only onhuman ingenuity, not physical constraints [11–13]. Thistechnological optimism has been met with a great deal ofcriticism during the past few years. A number of physicalscientists, as well as a growing minority of economists,argue that this view overlooks the uniqueness and finitenessof oil resources.

In this paper, we use a result of Hotelling [14] to examinethe hypothesis that technology will mitigate the finitenessof petroleum reserves. Hotelling presented a formalanalysis of non-renewable resources and derived somebasic implications as to how technological investment in afinite resource affects the resource price, the extractionpath, and the time until depletion.1 Although the assump-tions Hotelling used may have been unrealistic he was ableto isolate the effect of each variable if the others are heldconstant. We use his basic model to analyze the productionpatterns of two mega oil fields: the Forties in the North Seaand the Yates field in Texas—fields that haveapplied advanced enhanced oil recovery (EOR) extractiontechnology—and examine the relationship between exhaus-tion rates and technological change. We find that thepattern of depletion in the Forties and Yates fieldsgenerally follows Hotelling’s predictions. Our resultsquestion the traditional economic view that technologycan easily overcome oil scarcity and they raise concernsabout the capacity of current mega-fields to meet future oildemand.

Section 2 presents the basic theoretical structure of theHotelling model of non-renewable resources. The Fortiesand Yates oilfield case-study results are reported in Section3. Section 4 expands on one result of the Hotelling model:the case of an increase in demand. Finally, Section 5presents the conclusions of our work.

1990 2010 2020

Effect of new technology on

path of exhaustion

Pro

duction

2. Hotelling revisited

It is only a small exaggeration to say that mostcontemporary resource economics is a footnote to Ho-telling’s paper ‘‘The Economics of Exhaustible Resources.’’[14]. Following Krautkraemer [16], the basic Hotellingequation can be written as

Maximize

Ze�dt½BðqðtÞ;SðtÞÞ � CðqðtÞ;SðtÞÞ�dt (1)

1Reserves are classified as proven, probable or possible. To be classified

as ‘‘proven’’ the degree of certainty must be 90% [15].

subject to

_SðtÞ ¼ �qðtÞ;SðtÞX0; qðtÞX0;Sð0Þ ¼ S0; (2)

where B[q(t), S(t)] represents gross benefits, C[q(t),S(t)] arethe extraction costs; S is the remaining stock, d is the rateof discount, and q(t) is the time path for resourceextraction that maximizes the present value of the streamof net benefits from extraction. Note that this key equationfor resource policy only contains discounted monetaryunits. All information about the physical resource basecomes through the price system.It can be seen from Eq. (1) that any new technology that

lowers extraction costs, other things being equal, willincrease the net present benefits of extraction. The effect ofthis on the resource price and the time path of extractionare shown in Fig. 1. A technological improvement thatreduces extraction costs (from C to C0) results in a lowerresource price (the decline from P to P0) at the beginning ofthe time path of exhaustion and a steeper rise in price at theend, and it decreases the time left until exhaustion. Weargue below that Fig. 1 approximates what is actuallyhappening in the representative oil fields we examined.

Time

Fig. 2. Technological advance masks impending production declines.

ARTICLE IN PRESS

1987

EOR

Technology

0

5

10

15

20

25

30

35

Annual P

roduction (

10

6 M

3 p

er

Year)

198519801975 1990 1995 2000

Fig. 3. Historical pattern of production in the Forties oilfield (UK). Data

Source: UK Department of Trade and Industry.

2An oil platform is a large structure used to house workers and

machinery needed to drill and then produce oil and natural gas in the

J. Gowdy, R. Julia / Energy 32 (2007) 1448–14541450

If Fig. 1 is an accurate description of an exhaustibleresource, then its exhaustion path will show a patterndepicted in Fig. 2 below. Technological change increasesproduction for a while but then the path to exhaustionbecomes steeper.

To the extent that technological advance increasespresent output at the expense of future output this impliesthat prices may be misleading indicators of future resourcescarcity. Today’s lower prices may be the result ofextracting more of the future’s resource supply. This isessentially the claim currently being made by a number ofprominent energy researchers [17,18], as well as popularwriters on the subject of ‘‘peak oil’’ [19]. They argue thatthe approaching peak of world oil production is beingmasked by great improvements in extraction technologysince the 1990s and that these technological improvementswill make the decline steeper and the price rise more rapidin the years to come. This view is summarized by Zittel [20]in the conclusion to a study of North Sea oil production:

This analysis shows how false it is to base our confidenceon oil reserves being sufficient to maintain, let aloneincrease current production levels. Once the peakproduction of large fields is passed, the situation mayswitch very fast. According to our experience, individualoil companies are tempted to keep a high productionrate as long as possible, instead of planning to smooththe future decline. Therefore, it might be feared thatfuture decline will be the steeper, the longer the world’soil companies try to hold high production levels.

This is essentially the same point Hotelling made in hisclassic paper. Is there compelling empirical evidencesupporting this point of view?

3. Is extraction technology masking scarcity? Evidence from

the North Sea and West Texas

In this section, we present the production paths thatfollowed in the North Sea Forties and the West TexasYates mega oilfields using historical production datareported by the United Kingdom Department of Tradeand Industry and the Railroad Commission of Texas,respectively. We purposely focus on these two well-surveyed and well-documented individual fields—ratherthan more aggregate examples—so as to acknowledge thestrict scarcity of the resource, as assumed in Hotelling’sframework.

The Forties and the Yates have been subject during theirproductive years to the application of enhanced oilrecovery (EOR) technologies. EOR measures (also calledsecondary and tertiary measures) are often cited as a meansof significantly increasing future recoverable reserves of oilby mobilizing resources and enhancing the recovery of oiland gas in place. EOR techniques include boostingextraction rates by better drainage of the oil in place:when the pressure in the oil field has decreased to a levelwhere the viscosity of oil dominates the production profile

leading to lower extraction rates, pressure can be raisedartificially by water or gas injection, or by reducing theviscosity with the injection of steam or a chemicalsurfactant [21]. EOR technologies are widely used and dohave the potential to enhance oil production. However, wepresent evidence that these measures increase the produc-tion rates only temporarily; apparently EOR reassuresmake it possible to extract the oil faster for a few years, butpartially at least at the cost of steeper subsequent declinerates [22].

3.1. The North Sea Forties oilfield

The Forties oilfield (named Forties after the sea area inwhich it lies) is the first and largest oilfield in the UK NorthSea; it was discovered in 1970 and rapidly became thesymbol of Britain’s North Sea oil boom. The time path ofproduction in mega-fields such as this one shows acharacteristic shape, slightly different from the classicbell-shaped curve shown in Fig. 2. Oil companies usuallytry to recover high initial investments as fast as possible byinitially producing the fields at high rate. Production isnormally capped at the maximum flow rate that can bekept given the pipeline system’s capacity and the technol-ogy in place, generating a plateau for a while. Eventually adecline in production occurs, driven by the decrease ofpressure in the field and rising water levels.Fig. 3 illustrates the historical production pattern of the

Forties, and Table 1 shows the change in annualproduction rates (extraction rates) for selected years. Thefield began production in 1975, and within 3 years, itreached a plateau at 500,000 barrels per day. This plateaulasted for about 3 more years before the decline started in1981. During the period 1981–1986 yearly productiondeclined at an average rate of 6.4%.In an attempt to boost production and recover produc-

tion rates, in 1986 an additional (fifth) oil platform2 was

ARTICLE IN PRESS

Table 1

Evolution of extraction rates in the Forties (UK) in selected years

Selected Years Average change in extraction per year

106M3 %

1975–1978 9.45 473.25

1979–1980 0.84 0.29

1981–1986 �1.60 �6.40

1987 �0.69 �3.54

1988–1989 �3.37 �19.32

1990–2003 �0.70 �10.44

1987

y = -0.1123x + 5E+07

R2 = 0.9688

0

5

10

15

20

25

30

35

0 50 100 150 200 250 300 350 400 450

Cummulative Production (106 M3)

Annual P

roduction (

10

6 M

3 p

er

Year)

Fig. 4. Pattern of cumulative production of the Forties oilfield (UK).

Data Source: UK Department of Trade and Industry.

0

5

10

15

20

25

30

35

40

45

50

1927 1937 1948 1958 1968 1978 1988 1998

Annual P

roduction 1

06 B

arr

els

per

Year

Fig. 5. Historical pattern of production of the Yates oilfield. Data Source:

Railroad Commission of Texas.

J. Gowdy, R. Julia / Energy 32 (2007) 1448–1454 1451

built, and EOR measures were implemented in 1987 byinjecting gas (CO2) into the reservoir. EOR measures andthe additional platform did reduce the decline in produc-tion to about half of the previous average (from 6.4% to3.5%). However, during the 2 following years, annualproduction dropped sharply, to an average rate of declineof 19.32%. From then on, yearly production in the Fortiesdeclined at an average rate of 10.44% (see Table 1).Technology arrested the decline in production for a year,from 1987 to 1988, but then the decline resumed at a ratesteeper than before the additional investments in technol-ogy took place.

Our analysis is constrained by the fact that there is nocontrol case available to compare what the rates ofdepletion might have been without the application ofEOR technology. Hence, it is impossible to determineprecisely the contribution of EOR measures to the ultimaterecovery of the field. However, for offshore mega-fields likethe Forties, plotting annual production versus cumulativeproduction shows a characteristic behavior: as soon as the

(footnote continued)

ocean. Depending on the circumstances, the platform may be attached to

the ocean floor, consist of an artificial island, or be floating.

plateau has passed, the production decline is almost linear.Hence, the extrapolation of that decline line to the x-axispermits a good estimate of the field’s ultimate recovery [23].Fig. 4 shows annual production in the Forties’ field as afunction of cumulative production, with a straight declinetrend after the plateau.This pattern implies that the additional platform and gas

injection temporarily increased production without greatlyaffecting the field’s ultimate total recovery. The figuresuggests that once the field reached its peak productionlevel, EOR technology only slightly altered the path ofdepletion and the level of ultimate recovery, contradictingthe widely promoted view that technology has the potentialof increasing recoverable oil reserves enough to mitigateabsolute scarcity.3

3.2. The West Texas Yates oilfield

Another example of the effect of EOR measuresoccurred in the Yates oilfield—US 7th largest oilfield—located in Pecos County, West Texas. The historicalpattern of production of the Yates is quite different fromthe pattern observed in the Forties’ (see Fig. 5).The Yates field was discovered in 1926, and by 1927

reached a daily average production of 9099 barrels of oil.The field showed an early production peak in 1929, with aproduction level of 41 million barrels of oil. Rates startedto decline in 1930, and by 1941 had dropped to under sixmillion barrels. Temporary measures such as observance ofspecial rules with regards to water production in the fieldsallowed rates to increase again—reaching a new lower peakof 18 million barrels in 1948. After this year however, thefield experienced a continuous decline and the RailroadCommission of Texas allowed the injection of salt waterand gas into producing formations. These early measures

3A very similar pattern of production and depletion is found in another

well-surveyed offshore mega-field: Prudhoe Bay, located in Alaska, United

States.

ARTICLE IN PRESSJ. Gowdy, R. Julia / Energy 32 (2007) 1448–14541452

maintained pressure and retarded water encroachment,thus exposing a greater area of the reservoir to gravitydrainage. By the year 1979, production in the Yates hadmore than doubled.

In 1984, however, the field started to show signs ofdecline again, and rates of production decreased continu-ously during the following 8 years. New EOR measureswere implemented at that time in an attempt to recover,once again, extraction rates: reported is the application ofwater flooding, polymer injection and carbon dioxideflooding in 1993. These last attempts were successful forabout 4 years, but in 1997 production not only started tofall again, but it did so at rates much higher than the onesaveraged in the period 1985–1993 [21].

Fig. 6 shows the production of the Yates field during theperiod 1984–2002, and Table 2 shows the average rates ofdepletion that the Yates field showed before and after theEOR technology was applied. By 2002, production rateswere well below the ones reported prior to the implementa-tion of the EOR measures. Again, investment in technol-ogy temporarily recovered production rates, but it did so atthe expense of a steeper rate of decline in later years.

1993

y = -1E+06x + 3E+09

y = -3E+06x + 6E+09

0

5

10

15

20

25

1988 1990 1992 1994 1996 1998 2000 2002 2004Pro

duction p

er

Year

10

6 B

arr

els

per

Year

EOR Technology

Fig. 6. Pattern of production of the Yates oilfield during the period

1998–2002.

Table 2

Evolution of extraction rates in the Yates (Texas, USA) in selected years

Selected Years Change in extraction per year

106 barrels per year %

1991 �1.86 �8.49

1992 �1.90 �9.51

1993 �0.47 �2.60

1994 1.27 7.21

1995 0.96 5.07

1996 0.24 1.20

1997 0.42 2.11

1998 �1.70 �8.29

1999 �1.86 �22.95

2000 �3.38 �23.30

2001 �1.86 �29.20

2002 �1.17 �14.84

4. Demand, royalties, and geopolitics: Hotelling once again

We have argued that the evidence indicates thattechnological change increases production for a while butthen the path to exhaustion becomes steeper than it wouldhave been without the new technology. Ceteris paribus,technological change increases resource supply and de-creases the resource price for a while but results in sharplyhigher prices in later periods because the resource isexhausted faster than it would have been without the newtechnology. This implies that current prices are misleadingindicators of future resource scarcity: today’s lower pricesare the result of extracting more of the future’s resourcesupply. Furthermore, increasing demand for petroleum, asshown below, may complicate the worldwide situation.The case of increased demand and enhanced recovery

(resulting in a fall in production costs) can also be analyzedusing Hotelling’s framework (see Fig. 7).An improvement in recovery technology shifts the Hotelling

price path downward and rotates it to the left. An increase indemand shifts the price path up and to the left. The combinedeffect on price is that EOR offsets the price effect of increaseddemand (partially or completely) but both effects reduce thetime until resource exhaustion. In Fig. 6, EOR reduces thetime to exhaustion from T to T0 and increased demand fromT0 to T00. The implication of a possible sharp increase indemand for petroleum is that, even with the best technologiesin place, we may face global oil exhaustion much sooner thanwhat it is customarily anticipated.Because of the rapidly growing international demand for

petroleum—much of it coming from the United States andEurope, but a rapidly increasing share from China, India,and other developing nations—the world’s requirementfor petroleum is projected to increase considerably. IfHotelling’s predictions follow, then the world mayexperience a much steeper rise in prices and the time ofexhaustion may come sooner than expected. The casemakes compelling the need for policies to encouragealternative sources of energy and conservation, and

Time

Price

Original Price Path

C

C'

A Fall

In Cost

TT'

Effect of an increase in

technology resulting in

a fall in extraction costs

P

P'

Fall in C plus increase in demand

T"

P"

Fig. 7. The effects of EOR technology and increasing demand.

ARTICLE IN PRESSJ. Gowdy, R. Julia / Energy 32 (2007) 1448–1454 1453

certainly the need of reconstructing a world economywhose political institutions and physical infrastructures arenow based on the availability of cheap oil [24].

5. Discussion and conclusion

Petroleum resources are not infinite on our planet and inthe coming decades increasing pressure will be placed onthe ability of oil to supply energy for the world’s growingeconomies. Technological progress has been posed as themajor contributor in the coming years to the mobilizationof new reserves and increasing the lifetime of presentreserves. In this paper, we examined this hypothesis andshowed evidence from two well-documented mega-oil-fields—the Forties in the North Sea and the Yates in WestTexas—that support the counter-argument that technol-ogy, rather than substantially increasing the lifetime ofpresent reserves, may only mask increasing absolutescarcity, in accordance with Hotelling’s theoretical predic-tions. Technology does have the effect of increasingextraction initially and makes it possible to extract moreof the petroleum in a given field, but it does so at the cost ofincreasing the rate of depletion—and decreasing thelifetime of the resource. Contrary to standard economicassumptions, temporary low prices may be misleadingindicators of the ability of the economy to adjust toresource exhaustion. The Classical economists were awareof this but contemporary welfare economics has all butabandoned concern with real, physical production pro-cesses in favor of abstract models of exchange in anidealized purely mathematical framework.

If our results are generalizable, they have seriousimplications for future world oil supply. If the pattern ofdepletion in the Forties and the Yates oilfields holds for themega-fields in the Middle East this does not bode well forfuture petroleum supply and prices. Saudi Arabia—theworld’s leading oil producer—will soon face the peaking ofits already mature giant fields. The case is made explicit bySimmons [15]: most of Saudi Arabia’s oil output isgenerated by a few giant fields, of which Ghawar—theworld’s largest—is the most prolific. While the Yates andForties fields at their peaks produced about 30 millionbarrels of oil a year, the Ghawar field has produced at rateof 10 million barrels per day. The giant Saudi fields areaging (they were first developed 40–50 years ago), and arealready subject to EOR measures (water injection andother secondary measures) to maintain high levels ofproduction and compensate for the drop in natural fieldpressure. If the pattern of exhaustion of these fields followsthe ones shown in the Forties and Yates, then we canexpect accelerated rates of depletion some time in the nearfuture. In addition, as time goes on, the ratio of water to oilin these underground fields will rise to the point wherefurther oil extraction will become difficult, if not impos-sible. According to Simmons [15] there is little reason toassume that future Saudi exploration will result in the

discovery of new fields large enough to replace those nowin decline.We have gathered and analyzed data for two of the three

regions in the world that do actually publicly reporthistorical production for individual fields—namely theUnited Kingdom, the United States and Norway. Indivi-dual oilfield estimates are confidential in most othercountries, and are available only at great expense fromcommercial databases of variable quality [25]. Althoughresults from two fields alone do not make by themselves asolid case that technology accelerates depletion, theyshould lead the way to a larger analysis. It is impossibleto predict with any degree of reliability what the futurerates of oil production are likely to be without firstunderstanding field-by-field depletion rates [26]. It becomescritical then to develop and release reserve estimates,particularly for the mega-fields in Saudi Arabia, if we are toevaluate where the future of oil production, upon which theworld’s economies now depend.Our study shows that details are important. It is not

enough to say that technology will call forth substitutes asoil is exhausted. The critical question is how rapidlysupplies are depleted and how rapidly quality substitutescan be made available at reasonable prices. EOR methodsmay just accelerate depletion of existing reserves OR theymay accelerate depletion and capture significant additionalreserves. In this paper, we make a tentative case that, fortwo large fields at least, the first possibility was the case.What happens in other fields and for the industry as awhole depends critically on the physical characteristics ofthe fields and the nature of new technology. Can newtechnology improve the recovery rate significantly beyondthe current 35% average? To our knowledge there is notechnology now on the horizon that would do that. As wemove down the depletion curve the energy return on energyinvested (EROI) [27] will decrease. This is important sinceon the way up the peak EROI acts as a push and on theway down acts as a pull on the economic feasibility ofrecovery. With its low extraction costs and flexibilitypetroleum is a unique source of energy far superior to anysuggested alternatives. The ‘‘peak oil’’ debate has becomepolarized between those who say ‘‘there is nothing to worryabout, technology will solve everything’’ and ‘‘we’re aboutto run out of oil and world’s economies will collapse.’’Either one of these statements may be true and the realisticanswer may not be in the middle. But the first step ingauging the implications of peak oil is to critically examinewhat has happened in the past and try to extrapolate theresults to the future.

Acknowledgments

We are grateful for the valuable comments provided byJason Benno, Jerry Gilbert, Charles Hall, John Hallock,Alexey Voinov, and Jack Zagar. We would like to thankthe Texas Railroad Commission for providing data about

ARTICLE IN PRESSJ. Gowdy, R. Julia / Energy 32 (2007) 1448–14541454

the Yates field. All conclusions and opinions are theresponsibilities of the authors.

References

[1] Barnett H, Morse C. Scarcity and growth: the economics of natural

resource availability. Baltimore: John’s Hopkins University Press;

1963.

[2] Hanley N, Shogren J, White B. Introduction to environmental

economics. Oxford: Oxford University Press; 2001.

[3] Fehr E, Fischbacher U. Why social preferences matter—the impact of

non-selfish motives on competition, cooperation and incentives. Econ

J 2002;112:C1–C33.

[4] Gintis H. Game theory evolving. Princeton, NJ: Princeton University

Press; 2000.

[5] Gowdy J. The revolution in welfare economics and its

implications for environmental valuation. Land Econ 2004;

80(2):239–57.

[6] Kahneman D. Maps of bounded rationality: psychology for

behavioral economics. Am Econ Rev 2003;93:1449–75.

[7] Cleveland C. Natural resource scarcity and economic growth revisited:

economic and biophysical perspectives. In: Costanza R, editor.

Ecological economics: the science and management of sustainability.

New York: Columbia University Press; 1991. p. 289–317.

[8] Norgaard R. Economic indicators of resource scarcity: a critical

essay. J Environ Econ Manage 1990;19:19–25.

[9] Alcott B. Jevons’ paradox. Ecol Econ 2005;54(1):9–21.

[10] Tainter J, Allen T, Little A, Hoekstra T. Resource transitions and

energy gain: contexts and organization. Conserv Ecol 2003;7(3) See

also http://www.consecol.org/vol7/iss3/art4.

[11] Dasgupta P. Bottlenecks: review of Jared Diamond’s book collapse.

London Rev Books 2005;19(May):21–2.

[12] Easterbrook G. There goes the neighborhood. New York Times, 30

January 2005.

[13] Simon J. The ultimate resource. Princeton, NJ: Princeton University

Press; 1981.

[14] Hotelling H. The theory of exhaustible resources. J Polit Econ 1931;

39:137–75.

[15] Simmons M. Twilight in the desert: the coming Saudi oil shock and

the world economy. NJ: Wiley; 2005.

[16] Krautkraemer J. Nonrenewable resource scarcity. J Econ Literature

1998;XXXVI:2065–107.

[17] Hall C, Tharakan P, Hallock J, Cleveland C, Jefferson M.

Hydrocarbons and the evolution of human culture. Nature 2003;

426:318–22.

[18] Simmons M. For an overview of the evidence for and negative

consequences of peak oil see: /http://www.simmonsco-intl.comS[19] Heinberg R. The party’s over: oil, war and the fate of industrial

societies. Canada: New Society Publishers; 2003.

[20] Zittel W. Analysis of the UK oil production. L-B-System Technik,

February 2001. See /http://www.energiekrise.de/e/articles/Analysis

_of_UK_oil_production.pdfS[21] Zittel W, Schindler J. International Summer School on the Politics

and Economics of Renewable Energy at Salzburg University, 2003.

See /http://www.Energiekrise.de/e/articles/International-Summer-

School-Salzburg-2002.pdfS[22] Campbell CJ, Laherrere JH. The end of cheap oil. Sci Am

1998;March:80–6.

[23] Smith CJ. Yates oilfield. Handbook of Texas Online. See: /http://

www.tsha.utexas.edu/handbook/online/articles/YY/doy1.htmlS[24] Hallocak J, Tharakan P, Hall C, Jefferson M, WuW. Forecasting the

limits to the availability and diversity of global conventional oil

supply. Energy 2004;29:1673–96.

[25] Laherrere JH. Forecasting production from discovery. In: Proceed-

ings of the fourth international workshop on oil and gas depletion,

Lisbon, 2005. See also: /http://www.cge.uevora.pt/aspo2005/ASPO

2005_Proceedings.pdfS[26] Simmons M. Depletion: the forgotten part of supply and demand.

See: /http://www/simmonsco-intl.com/files/41.pdfS[27] Hall C, Cleveland C, Kaufman R. Energy and resource quality: the

ecology of the economic process. Niwot, CO: University of Colorado

Press; 1992.

[28] Pearce D, Turner K. Economics of natural resources and the

environment. Baltimore: The Johns Hopkins Press; 1990.